
foreclosure-analysis
by breverdbidder
BrevardBidderAI Landing Page - Agentic AI Foreclosure Intelligence Platform
SKILL.md
name: foreclosure-analysis description: Expert skill for analyzing Florida foreclosure auction properties using The Everest Ascent™ 12-stage methodology. Use when analyzing properties, calculating max bids, researching liens, or making bid/skip decisions for Brevard County foreclosure auctions.
Foreclosure Analysis Skill
Expert methodology for analyzing foreclosure auction properties in Brevard County, Florida using The Everest Ascent™ systematic approach.
When to Use This Skill
- Analyzing foreclosure auction properties
- Calculating maximum bid amounts
- Researching title and lien priority
- Making BID/REVIEW/SKIP recommendations
- Generating property analysis reports
- Evaluating exit strategies (wholesale, retail, rental)
The Everest Ascent™ 12-Stage Pipeline
Stage 1: Discovery
- Scrape auction listings from RealForeclose (brevard.realforeclose.com)
- Extract: case number, property address, plaintiff, defendant, judgment amount, auction date
- Filter for in-person Titusville courthouse auctions vs online tax deed sales
- Initial data validation and deduplication
Stage 2: Scraping & Enrichment
Data Sources (Priority Order):
-
BCPAO API (gis.brevardfl.gov/gissrv/rest/services/Base_Map/Parcel_New_WKID2881/MapServer/5)
- Property details: lot size, living area, year built, zoning
- Photos: masterPhotoUrl field (format: https://www.bcpao.us/photos/{prefix}/{account}011.jpg)
- Tax values: just value, market value, assessed value
-
RealForeclose
- Case details, auction status, sale date
- Historical auction results
-
Zillow/Redfin (for residential properties)
- Beds, baths, square footage (BCPAO GIS lacks these)
- Recent sales comps
- Market trends
Critical: BCPAO provides property data but NOT beds/baths for residential. Must use Zillow/Redfin for complete residential analysis.
Stage 3: Title Search
- Search BECA (Brevard County Clerk) for recorded documents
- Identify all mortgages, liens, judgments in chronological order
- NEVER GUESS - only use actual recorded documents
Stage 4: Lien Priority Analysis
Florida Foreclosure Law - Lien Priority:
-
First Position Mortgage Foreclosure (most common):
- Senior mortgage forecloses
- Wipes out junior liens (2nd mortgages, HELOCs, mechanic's liens)
- Buyer gets clean title except for:
- Property taxes (always survive)
- HOA assessments recorded BEFORE the foreclosing mortgage
- Code enforcement liens (sometimes)
-
HOA Foreclosure (CRITICAL SCENARIO):
- HOA forecloses for unpaid assessments
- Senior mortgage SURVIVES
- Buyer inherits the senior mortgage debt
- AUTOMATIC DO_NOT_BID unless bid covers: HOA judgment + senior mortgage balance + your profit
- Detection: Plaintiff is HOA/condo association, mortgage still on title
-
Tax Certificate Foreclosure:
- County sells tax lien certificates
- After 2 years, certificate holder can foreclose
- Wipes out ALL liens including mortgages
- Very clean title for buyer
AcclaimWeb Search Pattern:
1. Search property address
2. Download all mortgages (OR book/page references)
3. Download all liens, judgments, lis pendens
4. Check recording dates vs foreclosing lien date
5. Flag if HOA plaintiff + senior mortgage exists
Detection Logic:
- If plaintiff contains: "HOA", "Homeowners Association", "Condo Association", "POA"
- AND mortgage recorded BEFORE HOA lien
- → Senior mortgage survives
- → DO_NOT_BID
Stage 5: Tax Certificate Check
- Search RealTDM for outstanding tax certificates
- Check redemption status
- Tax certificates < 2 years: owner can still redeem
- Tax certificates ≥ 2 years: holder can foreclose
Stage 6: Demographics & Market Analysis
Census API Data:
- Median household income (target: $78-82K)
- Population trends
- Vacancy rates (target: 5-6%)
- Age demographics
Optimal Zip Codes (ML-validated):
- 32937 Satellite Beach (income: $82K, vacancy: 5.2%)
- 32940 Melbourne/Viera (income: $80K, vacancy: 5.8%)
- 32953 Merritt Island (income: $78K, vacancy: 6.1%)
- 32903 Indialantic (income: $81K, vacancy: 5.4%)
Red Flags:
- High crime zip codes: 32901, 32904, 32935
- Declining population trends
- Vacancy rates > 8%
Stage 7: ML Probability Score
XGBoost Model (64.4% accuracy):
- Predicts: Third-party purchase probability, estimated sale price
- Features: judgment amount, property value, zip code, plaintiff type, time on market
- Output: Probability score 0-100%
Plaintiff Risk Profiles (28 tracked):
- Wells Fargo: 72% third-party purchase (aggressive bidders)
- Bank of America: 65% third-party purchase
- US Bank: 58% third-party purchase
- Local credit unions: 45% third-party purchase (bank often bids)
Stage 8: Maximum Bid Calculation
Formula:
Max Bid = (ARV × 70%) - Repairs - $10,000 - MIN($25,000, 15% × ARV)
Components:
- ARV (After Repair Value): Zillow/Redfin comps, BCPAO market value
- Repairs: Estimate from photos, property age, condition
- $0-5K: Good condition, cosmetic only
- $10-25K: Moderate rehab (kitchen, baths, flooring)
- $30-50K: Major rehab (roof, HVAC, structural)
- $60K+: Gut rehab or structural issues
- $10K buffer: Holding costs, closing, surprises
- MIN($25K, 15% ARV): Profit margin
Example:
ARV: $350,000
Repairs: $20,000
Max Bid = ($350,000 × 0.70) - $20,000 - $10,000 - MIN($25,000, $52,500)
Max Bid = $245,000 - $20,000 - $10,000 - $25,000
Max Bid = $190,000
Stage 9: Decision Logic
Bid/Judgment Ratio:
Ratio = Max Bid ÷ Judgment Amount
Decision Rules:
- Ratio ≥ 0.75 (75%): BID - Strong deal, bid up to max
- Ratio 0.60-0.74 (60-74%): REVIEW - Marginal deal, need more analysis
- Ratio < 0.60 (<60%): SKIP - Insufficient margin
Additional BID Criteria:
- Clean title (no HOA foreclosure with senior mortgage)
- Repairs < 30% of ARV
- Property in target zip code
- Exit strategy viable (flip, rental, wholesale)
Additional SKIP Criteria:
- HOA foreclosure + senior mortgage survives
- Code violations > $10K
- Structural issues (foundation, severe water damage)
- Environmental issues (mold, asbestos, underground tanks)
- Title clouds (unresolved liens, boundary disputes)
Stage 10: Report Generation
One-Page DOCX Format:
- Header: BidDeed.AI branding (navy #1E3A5F)
- Property Photo: BCPAO masterPhotoUrl
- Key Metrics:
- Address, case number, auction date
- Judgment amount, max bid, bid/judgment ratio
- ARV, repair estimate, profit margin
- ML probability score
- Decision Box:
- BID (green #E8F5E9)
- REVIEW (orange #FFF3E0)
- SKIP (red #FFEBEE)
- Exit Strategy: Wholesale, flip, rental with projected returns
- Lien Summary: Clean or issues flagged
- Risk Factors: HOA, code violations, structural issues
NEVER include:
- Property360 references
- Mariam Shapira references
- Everest Capital of Brevard LLC (use "Everest Capital USA" or "BidDeed.AI")
Stage 11: Disposition Tracking
- Record actual auction outcome
- Track: winning bid, buyer, bank credit bid
- Compare predicted vs actual results
- Update ML model with ground truth data
Stage 12: Archive & Learning
- Save all analysis to Supabase
auction_resultstable - Document lessons learned
- Update decision thresholds based on actual results
- Refine repair estimates from actual rehab costs
Data Sources & APIs
BCPAO API:
Base: https://gis.brevardfl.gov/gissrv/rest/services/Base_Map/Parcel_New_WKID2881/MapServer/5
Query: /query?where=PARCEL_ID='...'&outFields=*&f=json
Photo: https://www.bcpao.us/photos/{prefix}/{account}011.jpg
Census API:
Base: https://api.census.gov/data/2021/acs/acs5
Key: Public (no auth needed for basic queries)
AcclaimWeb:
- Manual search (no API)
- Search by: address, name, book/page
- Download PDFs of mortgages/liens
RealForeclose:
- Scrape auction listings
- Use async httpx, anti-detection headers
- Respect rate limits
Supabase:
Project: mocerqjnksmhcjzxrewo.supabase.co
Tables: auction_results, historical_auctions, insights
Critical Rules
- NEVER guess about liens - only use recorded documents from BECA
- ALWAYS check for HOA foreclosures - senior mortgage survival is deal killer
- Use ACTUAL property data - no fake comps or estimated values
- BCPAO lacks beds/baths - must use Zillow/Redfin for residential
- One-page reports only - BidDeed.AI branding, DOCX format
- Fair Housing compliance - NO race/ethnicity/familial status in analysis
- Conservative estimates - better to skip than overbid
Exit Strategies
Wholesale (Fastest):
- Target: 65-70% ARV
- Timeline: 7-30 days
- Profit: $10-25K
- Best for: Clean properties, quick capital return
Retail Flip (Highest Profit):
- Target: 70-75% ARV + repairs
- Timeline: 3-6 months
- Profit: $40-80K
- Best for: Nice neighborhoods, cosmetic repairs
Mid-Term Rental (Third Sword):
- Target zip codes: 32937, 32940, 32953, 32903
- Timeline: Hold 1-3 years
- Cash flow: $800-1,500/month
- Best for: Furnished rentals, traveling professionals
Common Pitfalls to Avoid
- Bidding on HOA foreclosures without checking senior mortgages
- Trusting BCPAO beds/baths for residential (often inaccurate/missing)
- Underestimating repairs (always add 20% buffer)
- Ignoring code violations (can delay closing/sale)
- Overbidding due to auction fever (stick to max bid)
- Skipping title search (hidden liens can destroy deal)
Pro Tips
- Attend auctions to observe bidding patterns
- Build relationships with title companies
- Pre-arrange financing for quick closes
- Visit properties before auction (exterior only)
- Track competitor bidding patterns
- Keep $50-100K liquid for quick acquisitions
Example Analysis Workflow
1. Property discovered in RealForeclose scrape
2. BCPAO API pull → property data + photo
3. Zillow scrape → beds/baths/comps (if residential)
4. AcclaimWeb search → title/lien documents
5. Lien priority analysis → HOA check
6. Census API → demographics
7. ML model → probability score
8. Max bid calculation → ARV × 70% formula
9. Decision logic → BID/REVIEW/SKIP
10. Generate DOCX report → BidDeed.AI branded
11. Save to Supabase → auction_results table
This skill encodes 10+ years of Brevard County foreclosure investing experience into a systematic, repeatable process that maximizes ROI while minimizing risk.
Score
Total Score
Based on repository quality metrics
SKILL.mdファイルが含まれている
ライセンスが設定されている
100文字以上の説明がある
GitHub Stars 100以上
3ヶ月以内に更新がある
10回以上フォークされている
オープンIssueが50未満
プログラミング言語が設定されている
1つ以上のタグが設定されている
Reviews
Reviews coming soon